AI PHYSICS - Atomic Structure

Transcription

1 AI PHYSICS - Atomic Structure Dr. Brian Strom. brianstrom999@ aol.com blog: Abstract: With the help of Artificial Intelligence (AI) and Deep Learning, the structure of the atom is computed from the results of established experiments on ionization energies and spectral emissions. The AI searches for trends and patterns in the ionization energy levels - when electrons are energized to escape from an atom. The AI first observes that the depth of the atomic Potential Energy Well is directly proportional to the number of protons in the nucleus, but is not dependent on the number of neutrons. The AI computes the energy levels for a multi-layered ball of electrons in a Potential Energy Well, and compares them to the ionization energy levels and electron depths. It identifies close similarities and proposes that electrons simply fill the three-dimensional atomic Potential Energy Well around the nucleus - looselypacked for the lighter elements, and more tightly-packed for the heavier elements. The AI concludes that electrons are much larger than we presently imagine. Introduction: Simple physics experiments have been conducted over the centuries with numerous theories to explain the observations. Certain theories have become dominant and, in the modern era, these fundamental beliefs generally go unchallenged. This paper reexamines some basic observations in physics and, with the help of Artificial Intelligence, proposes an alternative explanation for the structure of the atom. Einstein s General Theory of Relativity proposes the distortion of the fabric of space by an object, creating a Potential Energy Well. Ionization energies and spectral emissions suggest the atom is a Potential Energy Well having a small nucleus at the centre with numerous electrons surrounding the nucleus. Bohr s model proposes fixed electron orbits whilst Quantum theory proposes probability functions. Neither theory satisfactorily explains the detailed nature of ionization energies and spectral emissions. Method: The AI starts from first principles. It is simply given the mathematics of a Potential Energy Well and datasets for electron ionization and spectral line emissions. It is not given any existing theories or explanations.

2 Firstly, it analyses the ionization energies required to remove the deepest electron from the atom in the first six elements. The numbers are taken from the Compendium of Chemical Terminology [1]. (An example of the dataset is shown in Annex 1.) Element H He Li Be B C No of protons energy to remove an electron (ev) Figure 1a. IONIZATION: Table of energies to remove electrons. The AI identifies that the Ionization Energy of the deepest electron is proportional to the square of the number of protons in the nucleus. In diagramatic form: Figure 1b. IONIZATION: Energies to remove deepest electrons Element Hydrogen Helium Lithium Beryllium Boron Carbon No of Protons neutrons or 6 6 Well depth (ev) x x x x x x 13.6 Energy ratio = Energy ratio = Energy ratio = Energy ratio = Energy ratio = Note: PE well is not dependent 30 on number of neutrons Energy ratio = Figure 1b. IONIZATION: Histogram of energies to remove deepest electrons.

3 Note: The AI does not find any mathematical pattern for electron ionization energies in relation to the supposed number of neutrons in the atom. This suggests that neutrons whatever their properties - do not reside in the nucleus, where their mass would contribute to the nature of the nuclear Potential Energy Well. The AI continues by analysing the differing ionization energies for each level of electrons for the first twelve elements, where the electrons are ejected one-by-one: (The main excitation levels for Hydrogen and Helium are also shown.) Element H He Li Be B C N O Fl Ne Na Mg No of protons Energy (ev) Figure 2a. IONIZATION - Table of energies to remove electrons. 0 H He Li Be B C N O Fl Ne Na Mg Figure 2b. IONIZATION - Histogram of energies to remove electrons.

4 The AI uses the mathematical relationship that the depth of an electron in the atomic Potential Energy Well is proportional to the square root of the energy required to remove the electron: Element H He Li Be B C N O Fl Ne Na Mg depth proportional to square root of energy Figure 3a. IONIZATION Table of electron depths ( square root of energy) 0-5 H He Li Be B C N O Fl Ne Na Mg Figure 3b. IONIZATION Histogram of electron depths ( square root of energy)

5 The AI observes that the depth of the atomic Potential Energy Well is directly proportional to the number of protons in the nucleus. It divides the electron depths by the number of protons in the nucleus, to show the comparative depths, as though each nucleus contained one proton only: Element H He Li Be B C N O Fl Ne Na Mg depth of electron (per proton) Figure 4a. IONIZATION: Table of electron depths per proton. IONIZATION - electron depths per proton H He Li Be B C N O Fl Ne Na Mg Figure 4b. IONIZATION: Histogram of electron depths per proton.

6 The AI normalizes the electron depths to give a unitary comparison: Element H He Li Be B C N O Fl Ne Na Mg Electron depths per proton (normalized) Figure 5a. IONIZATION: Table of electron depths per proton normalized H He Li Be B C N O Fl Ne Na Mg Figure 5b. IONIZATION: Histogram of electron depths per proton normalized.

7 ATOMIC SPECTRA: The AI is also given the experimentally observed spectral emission line wavelengths from the official data published by the National Institute of Standards and Technology [2]. (An example of the Dataset is shown in Annex 2.) We believe these spectral emission lines occur when electrons fall into the atom, collide with another electron, and emit photons. The freqency (and energy) of the spectral emission is inversely proportional to the emission wavelength. The AI analysed the spectral lines for the first 12 elements, plus Uranium: H He Li Be B C N O F Ne Na Mg U Figure 6a. SPECTRA: energy of electron fall. The depths the electrons fall into the atomic Potential Energy Well are proportional to the square root of the emission energies:

8 H He Li Be B C N O F Ne Na Mg U Figure 6b. SPECTRA: depth of electron fall H He Li Be B C N O F Ne Na Mg U Figure 6c. SPECTRA: depth of electron fall - per proton.

9 ANALYSIS: The AI analyses the dataset of ionization energies for electrons being ejected from the atom, and concludes that the depth of the three-dimensional atomic Potential Energy Well is dependent on the number of protons in the atom. As proton numbers increase, the depth of the Potential Energy Well increases in direct proportion. Note: At least for the lighter elements, the characteristics of the atomic Potential Energy Well do not appear to be dependent on the number of neutrons in the atom. For the spectral emissions dataset, the AI envisions electrons falling into the threedimensional Potential Energy Well of the atom. The further an electron falls, the more energetic is its spectral emission. Both the ionization energy, and the spectral emission energy, are proportional to the square of the depth of the electron in the Potential Energy Well. Intelligent Computation: The AI computes the energy levels for a multi-layered ball of spherical objects in a 3-dimensional Potential Energy Well: The closest a sphere can sit next to the centre of the Potential Energy Well is adjacent to the centre, at sphere radius r. For 2 spheres sitting side-by-side, the distance from their centres to the centre of the Potential Energy Well will also be sphere radius 1r. For 3 identical spheres in a flat plane, the equilibrium position for each sphere will be for its centre to be 155 r from the centre of the Potential Energy Well (Figure 7). The AI computes the numerous energy levels as more spheres are added. The energy steps become smaller as the number of spheres increases.

10 Figure 7. Dimensions for three close-packed electrons. Identification of patterns: The AI compares the ionization electron depths and the mathematics for spheres in a Potential Energy Well and identifies the similarities. It concludes that electrons simply fill the three-dimensional atomic Potential Energy Well around the nucleus, layer by layer. The AI divides the electron depths by the number of protons in the nucleus to show the normalized depths the electrons fall. The AI places electron centres at each position. (Figure 8.) For the lighter elements, the electrons appear to be loosely-packed. For Hydrogen, it is relatively easy for an incoming electron to fall through the loosely-packed electrons to the lower levels in the atom, even to the lowest level alongside the nucleus. For the heavier atoms, the electrons are more tightly-packed, so electrons falling into the Potential Energy Well will travel through fewer layers of electrons before colliding with one of them. For the elements with more protons and a deeper Potential Energy Well, the second layer electron energy level is seen to become asymptotically closer to the r position.

11 Figure 8. The AI envisions electrons at each energy point. Symmetry: The underlying structure and symmetry of the nucleus will be different for each element, depending on the number of protons. Consequently, the symmetry of the ball of electrons surrounding the nucleus will be slightly different for each element. For heavier atoms with more protons, there will be more layers of electrons. For an atom with a nucleus having a symmetrical arrangement of protons, as in the noble gases, the electron layers also appear to be more symmetrical, requiring higher ionization energies to remove electrons. The force on the falling electron will be much greater for the heavier elements with more protons in the nucleus, but the distance the electron falls is shorter in the heavier atoms because the existing electrons are more tightly-packed.

12 CONCLUSIONS: The AI analyses the electron ionization and emission data and concludes that the depth of the atomic Potential Energy Well is dependent on the number of protons in the atom. At least for the lighter elements, the characteristics of the atomic Potential Energy Well do not appear to be dependent on the number of neutrons in the atom. The AI proposes that electrons simply fill the three-dimensional Potential Energy Well around the atomic nucleus, layer by layer, without the need for unexplained electron orbits as required for the Bohr atomic model. The AI envisions that electrons are larger than we presently imagine. The AI image of the Hydrogen atom is a small cluster of electrons surrounding a nucleus of one proton: Figure 9. Hydrogen atom. The AI image of the Carbon atom is a larger cluster of electrons surrounding a nucleus of 6 protons. The volume of the three-dimensional atomic Potential Energy Well is larger than for Hydrogen and, therefore, the number of electrons sitting in the Potential Energy Well is greater.

13 Figure 10. Carbon atom. Figure 11. Models of Atomic Structure.

14 =========================================== REFERENCES [1] Compendium of Chemical Terminology, 2nd ed. Compiled by A. D. McNaught and A. Wilkinson. Blackwell Scientific Publications, Oxford (1997). [2] National Institute of Standards and Technolgy - Basic Atomic Spectroscopic Data: ============================================ ANNEX 1. Ionization Energies in the Compendium of Chemical Terminology: Figure 11. Ionization: Compendium of Chemical Terminology

15 ANNEX 2. Spectral emission lines for Hydrogen in the National Institute of Standards and Technolgy - Basic Atomic Spectroscopic Data. HYDROGEN Spectral energy proportional Spectral to inverse of Intensity Wavelength wavelength Figure 12. Spectral emission lines for Hydrogen. ===================================================================